6G    

 

 

 

TDoc Tracking - Channel Coding

There is broad convergence on reusing NR as the baseline—LDPC for data and Polar for control—while studying targeted extensions to meet 6G KPIs for peak rate, latency, reliability, and efficiency.

 Several companies propose scaling LDPC by increasing maximum block/lifting sizes to cut TB segmentation loss and raise throughput, and even exploring a new high-throughput base graph to improve convergence and decoder area/energy efficiency.

 Others emphasize hardware reuse/structural compatibility across 5G and 6G so that encoder/decoder IP can serve both RATs, keeping migration costs low and MRSS-friendly.

 For Polar, proposals focus on enabling larger payloads and two-stage/segmented decoding to reduce blind-decoding power while maintaining BLER and FAR targets on control channels.

 At the same time, contributors call for a clear, shared evaluation framework—which KPIs and code-level metrics to optimize (BLER/latency/area/energy/flexibility)—before reopening major design choices.

 Peak data-rate assumptions ranging from ~50 Gbps up to 200 Gbps are discussed as possible stress points for code/decoder throughput, with a caution to avoid minor tweaks that don’t translate to meaningful system-level gains.

 Overall, the direction is to evolve, not replace: retain NR code families and proven machinery (IR-HARQ, QC structures), and introduce pragmatic, hardware-aware enhancements where 6G’s higher throughput and device diversity truly warrant them

Baseline & design principles

In 6G Radio (6GR), channel coding builds on the proven foundation of 5G NR. LDPC continues to serve the data channel, and Polar remains the main option for control. This is not only because of their strong error correction capability, high degree of parallelism, and efficiency close to the Shannon limit, but also because this approach allows existing NR infrastructure and silicon to be reused with minimal disruption.

The guiding principle is pragmatic evolution rather than replacement. Instead of reopening debates on entirely new code families, 6GR focuses on targeted enhancements to the current framework. This aligns with the new performance goals of 6G, such as ultra-high throughput, wider bandwidth, lower latency, tighter energy constraints, and more flexible control signaling.

LDPC improvements focus on faster convergence with lower iteration counts, larger lifting sizes, and scalable base graph structures to support higher throughput. Better performance under higher-order modulation and probabilistic shaping is also a key direction. Polar codes are being rethought to reduce blind decoding complexity, support larger payload sizes, and scale more efficiently with evolving control requirements.

By extending and refining the existing NR coding framework, 6GR can achieve meaningful performance gains while maintaining backward compatibility. This ensures that cost, efficiency, and interoperability are preserved, avoiding unnecessary complexity or marginal alternatives.

  • Reuse NR code families as baseline — Keep LDPC for data and Polar for control in 6GR, extending only where KPIs require.
  • Evolve, don’t replace — Prioritize pragmatic enhancements over radical changes that don’t yield system-level benefits.

Design Principles and Guidelines

  • Known Theoretical Limits
    • 5G LDPC and Polar codes already operate close to the Shannon limit; achieving further coding gains beyond a few tenths of a dB is extremely challenging without excessive complexity
    • Additional shaping gain of up to 1.53 dB remains the theoretical maximum from joint coding–modulation design; enhancements should consider probabilistic and geometric shaping rather than expecting large standalone coding gains
  • Learning from 5G
    • Recognize lessons from LTE → NR migration: lengthy code-family debates delayed progress, while pragmatic decisions (LDPC + Polar) enabled rapid ecosystem adoption R
    • Address known 5G limitations such as:
      • Polar payload size capped at 140 bits for DCI, limiting scheduling flexibility
      • Limited effectiveness of Polar distributed CRC for early termination and false alarm resilience
      • LDPC segmentation overhead for large transport blocks, causing error-propagation effects
  • Evolve, Don’t Replace
    • Focus on incremental, pragmatic enhancements to existing 5G codes rather than wholesale replacement
    • Preserve compatibility with 5G hardware to enable reuse of encoder/decoder blocks across NR and 6GR, lowering cost for dual-mode devices
  • Reuse NR Code Families as Baseline
    • Maintain LDPC for data channels and Polar for control channels in 6GR, building on their proven performance and efficiency
    • Avoid reopening debates on alternative code families unless system-level KPIs demonstrate clear and justified benefits

LDPC for data channels

In the evolution to 6GR, LDPC codes remain the natural choice for data channels. Their role as the default coding family in NR continues. The design direction focuses on scaling rather than replacing. Larger code-block sizes and extended lifting factors are expected to reduce segmentation overhead and improve finite-blocklength performance. This becomes important as transport block sizes grow in high-throughput scenarios.

NR’s two base graphs have shown strong versatility and efficiency. New or modified base graphs should only be introduced if they provide clear system-level benefits. This includes supporting 6G peak data rate targets without driving up hardware cost unnecessarily. At the same time, decoding efficiency remains a critical factor. The inherent parallelism of LDPC decoding, combined with the continuity of existing NR implementations, offers a solid foundation to meet 6G latency and throughput targets. It also keeps power and silicon budgets within practical limits.

This approach reflects a pragmatic evolution of NR LDPC. By scaling and refining where necessary instead of fragmenting the design, 6GR can reach new performance frontiers while preserving backward compatibility.

  • Scale code-block sizes — Consider larger blocks/liftings to reduce TB segmentation and increase throughput; explore new or modified base graphs if needed.
  • NR LDPC as default — Introduce an additional base graph only if higher peak-rate targets demonstrably demand it.
  • Hardware-aware decoding — Leverage parallelism and continuity from existing NR decoders to hit 6G latency/throughput goals.

Design Principles and Guidelines

  • Known Theoretical Limits
    • LDPC  codes already operate close to the Shannon limit; achieving further coding gains beyond a few tenths of a dB is extremely challenging without excessive complexity.
    • Additional shaping gain of up to 1.53 dB remains the theoretical maximum from joint coding–modulation design; enhancements should consider probabilistic and geometric shaping rather than expecting large standalone coding gains.
  • Learning from 5G
    • Recognize lessons from LTE → NR migration: lengthy code-family debates delayed progress, while pragmatic decisions (LDPC + Polar) enabled rapid ecosystem adoption.
    • Address known 5G limitations such as:
      • LDPC segmentation overhead for large transport blocks, causing error-propagation effects.
  • Scale Code-Block Sizes
    • Consider larger block sizes and extended lifting values to reduce segmentation overhead in large transport blocks and improve finite-blocklength performance.
    • Ensure scaling supports high-throughput operation (e.g., >50–200 Gbps peak rates) without excessive error propagation across multiple code blocks.
  • NR LDPC as Default
    • Use NR LDPC base graphs (BG1/BG2) as the primary baseline for 6GR, given their proven efficiency and compatibility.
    • Introduce new base graphs only where needed to meet peak-rate or low-latency demands, e.g., a High-Throughput Base Graph optimized for fast convergence and reduced complexity.
  • Hardware-Aware Decoding
    • Maintain quasi-cyclic structures and layered decoding to exploit parallelism and minimize latency.
    • Ensure decoder designs remain backward-compatible with 5G NR hardware, enabling reuse of encoder/decoder blocks for dual-mode operation.
    • Optimize for small-iteration convergence to support ultra-high throughput without linearly scaling silicon area.
  • Adaptation to Higher-Order Modulation
    • Improve base graph design for 256QAM and beyond, recognizing reliability variations across bit positions.
    • Leverage probabilistic and geometric shaping to capture shaping gain while balancing decoding complexity.
  • System Efficiency
    • Design LDPC enhancements with energy efficiency and area savings in mind, critical for dense 6G deployments.
    • Preserve encoding latency benefits from NR’s dual-diagonal and Raptor-like structures while extending them to larger lifting sizes.
  • Evaluation and Metrics
    • Use normalized iteration count and edge density per base graph as metrics to capture complexity–performance trade-offs.
    • Include throughput per area, energy per bit decoded, and scalability under FR3 bandwidths as additional benchmarks for 6GR.

Polar for control channels

In 6GR, Polar codes continue to serve as the natural foundation for control channels. Their role carries over from NR, where they were selected for their strong error correction capability at short blocklengths and their suitability for critical signaling. The main direction is to keep Polar as the default while refining its structure to meet new 6G requirements. This includes extending support for larger payloads and applying segmented or two-stage decoding to balance complexity, power consumption, and blind-decoding efficiency.

Pain points observed in 5G must be addressed. These include the limited scalability of DCI payloads beyond 140 bits, the weak effectiveness of distributed CRC for early termination, and the vulnerability to false alarms during blind decoding. Enhancements need to improve termination behavior, reduce decoding latency, and lower power consumption on the UE side. They must also strengthen reliability so that BLER and false alarm targets are met in denser and more demanding deployment environments.

This pragmatic refinement of Polar builds on its proven strengths. It ensures that control channel coding evolves in step with the broader 6G performance envelope without losing the efficiency and robustness established in NR.

  • Retain Polar, refine where needed — Enable larger payloads and segmented/two-stage decoding to manage power and blind-decoding complexity.
  • Resolve known pain points — Address DCI payload scaling and improve early-termination behavior while meeting BLER/FAR targets.

Design Principles and Guidelines

  • Known Theoretical Limits
    • Polar codes already operate close to the Shannon limit for short blocklengths; further coding gain is challenging without incurring significant complexity.
    • Performance improvements should therefore focus on reducing decoding latency, complexity, and false alarm rate rather than expecting large standalone coding gains.
  • Learning from 5G
    • Polar codes were effective for short control payloads in NR but showed limitations in scalability and robustness under practical blind decoding conditions.
    • Known issues include:
      • DCI payload size capped at 140 bits, limiting scheduling flexibility.
      • Distributed CRC design provides limited early termination, increasing UE decoding load.
      • False alarm risks with RNTI scrambling in dense deployments, threatening control channel reliability.
  • Retain Polar, Refine Where Needed
    • Keep Polar as the default for control channels, leveraging existing ecosystem maturity and hardware support.
    • Enable larger payloads through segmented or two-stage decoding while maintaining low power consumption and efficient blind detection.
    • Introduce design refinements that allow scalable payload sizes, supporting broader 6G scheduling needs.
  • Resolve Known Pain Points
    • Improve early termination mechanisms to reduce average decoding complexity and latency for UEs.
    • Enhance scalability of distributed CRC or introduce flexible alternatives to support larger payload sizes without degrading BLER/FAR targets.
    • Mitigate false alarm risks in RNTI-based scrambling by strengthening code construction and validation methods.
  • System Efficiency
    • Focus refinements on reducing UE power consumption during blind decoding, which dominates control channel energy cost.
    • Ensure proposed changes are lightweight and compatible with existing SCL decoding hardware to minimize additional implementation burden.
  • Evaluation and Metrics
    • Evaluate proposed enhancements against BLER and false alarm rate (FAR) requirements under realistic blind decoding scenarios.
    • Use metrics such as decoding latency, average iterations until termination, and computational complexity savings to quantify gains.
    • Ensure refinements maintain robustness across diverse payload sizes, channel conditions, and deployment densities.

Joint coding & modulation (JCM) / shaping

In 6GR, joint coding and modulation (JCM) becomes a key path to push spectral efficiency closer to theoretical limits without introducing new code families. LDPC and Polar already operate near capacity under conventional assumptions. However, additional shaping gains of up to about 1.5 dB can be achieved by aligning coded bits with the unequal reliability levels in higher-order QAM constellations. Probabilistic and geometric shaping provide a practical way to realize these gains.

The goal is not to disrupt the existing PHY framework with radical changes. Instead, shaping should be applied in a pragmatic way that keeps complexity under control and remains compatible with current LDPC and Polar structures. At the same time, 6G opens new opportunities to use side information more intelligently. Channel quality, modulation order, and receiver context can guide coding and mapping strategies. This allows further performance improvements without major architectural shifts.

By embedding shaping and side-information exploitation within the established LDPC and Polar frameworks, JCM can enhance both efficiency and robustness. This approach ensures forward compatibility and practical deployment in 6GR.

  • Pursue shaping gains — Study joint coding–modulation approaches to harvest additional efficiency without excessive overhead.
  • Use side information wisely — Consider designs that exploit transmitter/receiver side information within conventional frameworks.

Design Principles and Guidelines

  • Known Theoretical Limits
    • Standalone coding gains beyond a few tenths of a dB are difficult to achieve, as LDPC and Polar codes already operate close to capacity.
    • The maximum shaping gain from joint coding–modulation approaches is bounded at approximately 1.53 dB, setting a clear target for practical designs.
  • Learning from 5G
    • NR LDPC was optimized mainly for QPSK, with less emphasis on higher-order modulation where bit reliability varies across symbol positions.
    • Experience from 5G shows that constellation shaping and coding–modulation alignment can provide efficiency improvements without redesigning the entire coding framework.
  • Pursue Shaping Gains
    • Investigate probabilistic and geometric shaping techniques integrated with LDPC and Polar coding to capture shaping gain efficiently.
    • Design enhancements should target improved robustness for higher-order QAM (e.g., 256QAM and beyond) where unequal bit reliability can be exploited.
    • Balance shaping gain with decoder complexity to ensure power and latency targets remain achievable.
  • Use Side Information Wisely
    • Explore coding–modulation designs that leverage available side information such as channel state, modulation order, or reliability levels.
    • Ensure side-information usage is compatible with existing PHY procedures and does not introduce excessive overhead or signaling burden.
    • Allow flexible adaptation across device classes, enabling both eMBB and IoT devices to benefit within the same coding framework.
  • System Efficiency
    • Target shaping methods that reduce energy per bit while maintaining low hardware cost and decoding latency.
    • Prioritize solutions that can be embedded into existing LDPC/Polar implementations to maximize reuse of NR-era infrastructure.
  • Evaluation and Metrics
    • Assess shaping and JCM approaches based on achievable SNR gain, complexity per iteration, and energy efficiency improvements.
    • Benchmark against BLER and throughput performance across modulation orders and fading conditions.
    • Include implementation trade-offs such as decoder area and memory footprint in evaluation criteria.

KPIs, evaluation metrics & methodology

In moving toward 6GR channel coding design, establishing clear KPIs, evaluation metrics, and methodology is critical. This provides a common basis for comparing candidate solutions and making down-selection decisions. Agreement on the main drivers—peak data rate, latency, reliability, and area or energy efficiency—needs to come early. Consistency across vendors and deployment scenarios depends on this shared foundation.

NR already offers a solid reference point. Its KPI definitions, simulation assumptions, and measurement methods for BLER, complexity, and latency are well understood and widely adopted. For 6G, these frameworks should be extended where necessary. The scope must reflect more ambitious goals, such as extreme peak data rates, wideband operation in FR3, and tighter control-channel performance under blind decoding conditions.

By anchoring evaluations in NR-style metrics and adapting them to 6G’s broader performance targets, the study process can remain structured and transparent. This allows enhancements to LDPC and Polar to be judged fairly in terms of both performance and implementation feasibility, maintaining a balance between gain, complexity, area, and power efficiency.

  • Align on drivers & metrics — Agree on peak rate, latency, reliability, area/energy KPIs and how to evaluate them before down-selection.
  • Start from NR KPI sets — Use NR-style measurements as a baseline, extending for 6G where justified.
  • Stress targets in scope — Consider very high peak data rates and tight control-channel targets when sizing codes and decoders.

Design Principles and Guidelines

  • Align on Drivers & Metrics
    • Define key performance drivers—peak data rate, latency, reliability, and area/energy efficiency—before evaluating candidate solutions.
    • Ensure all proposals are compared under a consistent evaluation framework to avoid fragmented decision-making.
  • Start from NR KPI Sets
    • Use NR-style KPI definitions (e.g., BLER, average number of iterations, decoding latency) as the baseline for evaluation.
    • Extend and refine NR metrics only where new 6G requirements clearly justify additional stress conditions.
  • Stress Targets in Scope
    • Explicitly consider very high peak data rate targets (tens to hundreds of Gbps) in throughput and area-efficiency analysis.
    • Include strict control-channel KPIs such as false alarm rate (FAR), early termination effectiveness, and BLER thresholds for critical signaling.
    • Ensure both data and control channel evaluations account for low-latency and high-reliability scenarios (e.g., HRLLC, immersive communication).
  • Comprehensive Complexity Metrics
    • Measure decoding complexity using normalized iterations, edge density, and memory footprint of base graphs or polar structures.
    • Track area and energy per bit decoded, providing a realistic view of implementation efficiency across device classes.
  • Unified Evaluation Methodology
    • Apply consistent assumptions for channel models (e.g., AWGN, fading), modulation formats, and block lengths.
    • Use standardized test cases across eMBB, IoT, and HRLLC to ensure solutions scale across device types and use cases.
    • Include both simulation-based performance metrics and hardware-aware efficiency estimates in the evaluation framework.

Implementation practicality & migration

In 6GR channel coding design, implementation practicality and smooth migration from NR are as important as performance itself. The global ecosystem has already invested heavily in 5G hardware, IP blocks, and multi-RAT spectrum sharing. Maximizing reuse of these assets is essential to keep costs under control and speed up deployment.

This requires LDPC and Polar enhancements to stay structurally compatible with existing decoder architectures. Any new features must integrate cleanly with MRSS frameworks, avoiding disruptive redesign. The priority should be on system-level gains such as higher throughput, lower latency, or reduced power consumption, not on marginal refinements that increase complexity without real benefits.

By keeping hardware reuse and practical improvements at the center of the design, 6GR coding evolution can achieve stepwise efficiency gains. This ensures smooth migration, avoids fragmentation, and minimizes deployment overhead.

  • Maximize hardware reuse — Keep designs MRSS-friendly and compatible with existing 5G IP to minimize migration cost.
  • Focus on tangible wins — Avoid tweaks that don’t translate into real improvements in throughput, latency, or energy.

Design Principles and Guidelines

  • Maximize Hardware Reuse
    • Ensure LDPC and Polar refinements remain structurally compatible with existing NR encoders and decoders.
    • Design with MRSS (Multi-RAT Spectrum Sharing) friendliness in mind to allow dual 5G/6G operation without costly hardware redesigns.
    • Preserve quasi-cyclic and layered decoding structures to leverage proven NR implementation efficiency.
  • Focus on Tangible Wins
    • Pursue enhancements that translate into measurable gains in throughput, latency, or energy efficiency.
    • Avoid complexity-heavy modifications that fail to deliver system-level performance improvements.
    • Prioritize scalable solutions that benefit both high-performance eMBB and resource-constrained IoT devices.
  • Migration Efficiency
    • Enable smooth transition from 5G to 6G by designing codes and decoders that can be reconfigured rather than replaced.
    • Support dual-mode devices by allowing shared encoder/decoder hardware across NR and 6GR.
    • Minimize additional silicon area, memory footprint, and logic overhead to lower cost of adoption.
  • System Integration
    • Align coding enhancements with broader PHY/RAN requirements such as MRSS coexistence, high-order modulation, and ultra-low latency use cases.
    • Ensure changes are implementation-friendly for both gNB and UE hardware, keeping energy and area efficiency central to design.

Study scope alignment (SID guidance)

In guiding the 6GR study, channel coding must be treated as part of a larger physical-layer design effort. It needs to align closely with waveform evolution, numerology, MIMO strategies, duplexing approaches, and initial access procedures. The Study Item already identifies LDPC for data and Polar for control as the baselines. Coding discussions should build on this foundation rather than revisiting fundamental choices.

Early clarity of scope is just as critical. Requirements, evaluation metrics, and the specific areas where enhancements are needed should be defined upfront. This keeps the work focused on closing real performance gaps and achieving agreed KPI targets. It prevents the study from drifting into unfocused or overlapping discussions.

By placing coding within the broader 6G system study and fixing its scope early, the process becomes more disciplined and coherent. This minimizes fragmentation, reduces unnecessary churn, and ensures that any proposed refinements contribute real value to the complete air-interface design.

  • Place coding within the larger study — Treat channel coding alongside waveform, numerology, MIMO, duplexing, and initial access, with LDPC/Polar as stated baselines.
  • Clarify scope early — Lock requirements, metrics, and the necessity of enhancements before re-opening major design choices.

Design Principles and Guidelines

  • Place Coding Within the Larger Study
    • Treat channel coding as one element of the broader 6GR physical-layer design alongside waveform, numerology, MIMO, duplexing, and initial access.
    • Ensure coding enhancements are evaluated in context with other PHY components to maintain balance in complexity, performance, and efficiency.
    • Respect the stated baseline: LDPC for data channels and Polar for control channels, as established in the study item guidance.
  • Clarify Scope Early
    • Lock key requirements and evaluation metrics before initiating enhancement studies to avoid repeated re-scoping.
    • Identify where enhancements are truly necessary to meet new 6G KPIs, rather than revisiting well-established design choices.
    • Prevent unnecessary debates on code-family selection by focusing efforts on targeted improvements to LDPC and Polar.
  • Alignment With Study Item (SID) Guidance
    • Follow the scope and priorities set by the approved 6GR Study Item, ensuring coding contributions are consistent with agreed objectives.
    • Coordinate with related study areas (e.g., MRSS, control-plane latency, FR3 bandwidths) to keep coding enhancements interoperable within the overall 6G framework.
    • Maintain flexibility to adjust within the SID boundaries but avoid reopening settled design directions without compelling evidence of system-level benefit.

Additional requirements & operating ranges

For 6GR channel coding, the requirements extend far beyond simple throughput scaling. The design must support a broader range of operating conditions to meet diverse service needs. Ultra-reliable and low-latency scenarios demand that control channels and HARQ-protected data achieve extremely low block error rates under tight latency constraints. These guarantees must strengthen and extend what was already established in 5G NR.

At the same time, 6G must accommodate a much wider envelope of code configurations. It needs to support very small block lengths for IoT and uplink control signaling, as well as extremely large transport blocks made possible by wideband FR3 operation. Code rates must span a broad range, from low to high, while maintaining stable performance.

The key challenge is achieving this flexibility without driving up decoder complexity. Solutions must remain efficient, scalable, and practical, covering both high-performance devices and resource-constrained endpoints. This balance is essential to make 6GR coding a robust and deployable foundation across all service categories.

  • Reliability & latency — Support stringent HRLLC-like targets (e.g., very low BLER within short latency budgets) on control and robust HARQ for data.
  • Wide code envelopes — Cover very small to very large block lengths and broad code-rate ranges while keeping decoder complexity in check.

Design Principles and Guidelines

  • Reliability & Latency
    • Support stringent reliability targets, including BLER levels as low as 10⁻⁵ for critical control and HRLLC traffic.
    • Meet ultra-low latency requirements by ensuring encoding/decoding operations fit within short processing budgets.
    • Provide robust HARQ performance for data channels, maintaining reliability under high-load and wideband conditions.
  • Wide Code Envelopes
    • Cover a broad span of block lengths, from very small sizes for IoT/control signaling to very large transport blocks for high-throughput FR3 operation.
    • Support a wide range of code rates (from low-rate robust coding to high-rate spectrum-efficient modes) within a unified framework.
    • Maintain scalability across payload sizes without excessive segmentation or error-propagation effects.
  • Complexity Control
    • Ensure decoding complexity remains practical even under extreme blocklength and rate configurations.
    • Use hardware-aware designs (e.g., layered decoding, parallelization) to handle wide operating ranges without exponential cost increases.
    • Balance coding gain against implementation burden, prioritizing efficiency across diverse device classes.
  • Unified Framework
    • Design LDPC and Polar enhancements to flexibly cover these ranges without introducing multiple disjoint code families.
    • Preserve consistency in decoder architecture across small and large payloads to minimize migration complexity.

Reference